/*
* opl3_hw.h - Definitions of the OPL-3 registers
*
*
* Copyright (C) by Hannu Savolainen 1993-1997
*
* OSS/Free for Linux is distributed under the GNU GENERAL PUBLIC LICENSE (GPL)
* Version 2 (June 1991). See the "COPYING" file distributed with this software
* for more info.
*
*
* The OPL-3 mode is switched on by writing 0x01, to the offset 5
* of the right side.
*
* Another special register at the right side is at offset 4. It contains
* a bit mask defining which voices are used as 4 OP voices.
*
* The percussive mode is implemented in the left side only.
*
* With the above exceptions the both sides can be operated independently.
*
* A 4 OP voice can be created by setting the corresponding
* bit at offset 4 of the right side.
*
* For example setting the rightmost bit (0x01) changes the
* first voice on the right side to the 4 OP mode. The fourth
* voice is made inaccessible.
*
* If a voice is set to the 2 OP mode, it works like 2 OP modes
* of the original YM3812 (AdLib). In addition the voice can
* be connected the left, right or both stereo channels. It can
* even be left unconnected. This works with 4 OP voices also.
*
* The stereo connection bits are located in the FEEDBACK_CONNECTION
* register of the voice (0xC0-0xC8). In 4 OP voices these bits are
* in the second half of the voice.
*/
/*
* Register numbers for the global registers
*/
#define TEST_REGISTER 0x01
#define ENABLE_WAVE_SELECT 0x20
#define TIMER1_REGISTER 0x02
#define TIMER2_REGISTER 0x03
#define TIMER_CONTROL_REGISTER 0x04 /* Left side */
#define IRQ_RESET 0x80
#define TIMER1_MASK 0x40
#define TIMER2_MASK 0x20
#define TIMER1_START 0x01
#define TIMER2_START 0x02
#define CONNECTION_SELECT_REGISTER 0x04 /* Right side */
#define RIGHT_4OP_0 0x01
#define RIGHT_4OP_1 0x02
#define RIGHT_4OP_2 0x04
#define LEFT_4OP_0 0x08
#define LEFT_4OP_1 0x10
#define LEFT_4OP_2 0x20
#define OPL3_MODE_REGISTER 0x05 /* Right side */
#define OPL3_ENABLE 0x01
#define OPL4_ENABLE 0x02
#define KBD_SPLIT_REGISTER 0x08 /* Left side */
#define COMPOSITE_SINE_WAVE_MODE 0x80 /* Don't use with OPL-3? */
#define KEYBOARD_SPLIT 0x40
#define PERCOSSION_REGISTER 0xbd /* Left side only */
#define TREMOLO_DEPTH 0x80
#define VIBRATO_DEPTH 0x40
#define PERCOSSION_ENABLE 0x20
#define BASSDRUM_ON 0x10
#define SNAREDRUM_ON 0x08
#define TOMTOM_ON 0x04
#define CYMBAL_ON 0x02
#define HIHAT_ON 0x01
/*
* Offsets to the register banks for operators. To get the
* register number just add the operator offset to the bank offset
*
* AM/VIB/EG/KSR/Multiple (0x20 to 0x35)
*/
#define AM_VIB 0x20
#define TREMOLO_ON 0x80
#define VIBRATO_ON 0x40
#define SUSTAIN_ON 0x20
#define KSR 0x10 /* Key scaling rate */
#define MULTIPLE_MASK 0x0f /* Frequency multiplier */
/*
* KSL/Total level (0x40 to 0x55)
*/
#define KSL_LEVEL 0x40
#define KSL_MASK 0xc0 /* Envelope scaling bits */
#define TOTAL_LEVEL_MASK 0x3f /* Strength (volume) of OP */
/*
* Attack / Decay rate (0x60 to 0x75)
*/
#define ATTACK_DECAY 0x60
#define ATTACK_MASK 0xf0
#define DECAY_MASK 0x0f
/*
* Sustain level / Release rate (0x80 to 0x95)
*/
#define SUSTAIN_RELEASE 0x80
#define SUSTAIN_MASK 0xf0
#define RELEASE_MASK 0x0f
/*
* Wave select (0xE0 to 0xF5)
*/
#define WAVE_SELECT 0xe0
/*
* Offsets to the register banks for voices. Just add to the
* voice number to get the register number.
*
* F-Number low bits (0xA0 to 0xA8).
*/
#define FNUM_LOW 0xa0
/*
* F-number high bits / Key on / Block (octave) (0xB0 to 0xB8)
*/
#define KEYON_BLOCK 0xb0
#define KEYON_BIT 0x20
#define BLOCKNUM_MASK 0x1c
#define FNUM_HIGH_MASK 0x03
/*
* Feedback / Connection (0xc0 to 0xc8)
*
* These registers have two new bits when the OPL-3 mode
* is selected. These bits controls connecting the voice
* to the stereo channels. For 4 OP voices this bit is
* defined in the second half of the voice (add 3 to the
* register offset).
*
* For 4 OP voices the connection bit is used in the
* both halves (gives 4 ways to connect the operators).
*/
#define FEEDBACK_CONNECTION 0xc0
#define FEEDBACK_MASK 0x0e /* Valid just for 1st OP of a voice */
#define CONNECTION_BIT 0x01
/*
* In the 4 OP mode there is four possible configurations how the
* operators can be connected together (in 2 OP modes there is just
* AM or FM). The 4 OP connection mode is defined by the rightmost
* bit of the FEEDBACK_CONNECTION (0xC0-0xC8) on the both halves.
*
* First half Second half Mode
*
* +---+
* v |
* 0 0 >+-1-+--2--3--4-->
*
*
*
* +---+
* | |
* 0 1 >+-1-+--2-+
* |->
* >--3----4-+
*
* +---+
* | |
* 1 0 >+-1-+-----+
* |->
* >--2--3--4-+
*
* +---+
* | |
* 1 1 >+-1-+--+
* |
* >--2--3-+->
* |
* >--4----+
*/
#define STEREO_BITS 0x30 /* OPL-3 only */
#define VOICE_TO_LEFT 0x10
#define VOICE_TO_RIGHT 0x20
/*
* Definition table for the physical voices
*/
struct physical_voice_info {
unsigned char voice_num;
unsigned char voice_mode; /* 0=unavailable, 2=2 OP, 4=4 OP */
unsigned short ioaddr; /* I/O port (left or right side) */
unsigned char op[4]; /* Operator offsets */
};
/*
* There is 18 possible 2 OP voices
* (9 in the left and 9 in the right).
* The first OP is the modulator and 2nd is the carrier.
*
* The first three voices in the both sides may be connected
* with another voice to a 4 OP voice. For example voice 0
* can be connected with voice 3. The operators of voice 3 are
* used as operators 3 and 4 of the new 4 OP voice.
* In this case the 2 OP voice number 0 is the 'first half' and
* voice 3 is the second.
*/
#define USE_LEFT 0
#define USE_RIGHT 1
static struct physical_voice_info pv_map[18] =
{
/* No Mode Side OP1 OP2 OP3 OP4 */
/* --------------------------------------------------- */
{ 0, 2, USE_LEFT, {0x00, 0x03, 0x08, 0x0b}},
{ 1, 2, USE_LEFT, {0x01, 0x04, 0x09, 0x0c}},
{ 2, 2, USE_LEFT, {0x02, 0x05, 0x0a, 0x0d}},
{ 3, 2, USE_LEFT, {0x08, 0x0b, 0x00, 0x00}},
{ 4, 2, USE_LEFT, {0x09, 0x0c, 0x00, 0x00}},
{ 5, 2, USE_LEFT, {0x0a, 0x0d, 0x00, 0x00}},
{ 6, 2, USE_LEFT, {0x10, 0x13, 0x00, 0x00}}, /* Used by percussive voices */
{ 7, 2, USE_LEFT, {0x11, 0x14, 0x00, 0x00}}, /* if the percussive mode */
{ 8, 2, USE_LEFT, {0x12, 0x15, 0x00, 0x00}}, /* is selected */
{ 0, 2, USE_RIGHT, {0x00, 0x03, 0x08, 0x0b}},
{ 1, 2, USE_RIGHT, {0x01, 0x04, 0x09, 0x0c}},
{ 2, 2, USE_RIGHT, {0x02, 0x05, 0x0a, 0x0d}},
{ 3, 2, USE_RIGHT, {0x08, 0x0b, 0x00, 0x00}},
{ 4, 2, USE_RIGHT, {0x09, 0x0c, 0x00, 0x00}},
{ 5, 2, USE_RIGHT, {0x0a, 0x0d, 0x00, 0x00}},
{ 6, 2, USE_RIGHT, {0x10, 0x13, 0x00, 0x00}},
{ 7, 2, USE_RIGHT, {0x11, 0x14, 0x00, 0x00}},
{ 8, 2, USE_RIGHT, {0x12, 0x15, 0x00, 0x00}}
};
/*
* DMA buffer calls
*/
ETMODE, KD_TEXT):
intel_cleanup_plane_fb+0x5b/0xa0 [i915]
drm_atomic_helper_cleanup_planes+0x6f/0x90 [drm_kms_helper]
intel_atomic_commit_tail+0x749/0xfe0 [i915]
intel_atomic_commit+0x3cb/0x4f0 [i915]
drm_atomic_commit+0x4b/0x50 [drm]
restore_fbdev_mode+0x14c/0x2a0 [drm_kms_helper]
drm_fb_helper_restore_fbdev_mode_unlocked+0x34/0x80 [drm_kms_helper]
drm_fb_helper_set_par+0x2d/0x60 [drm_kms_helper]
intel_fbdev_set_par+0x18/0x70 [i915]
fb_set_var+0x236/0x460
fbcon_blank+0x30f/0x350
do_unblank_screen+0xd2/0x1a0
vt_ioctl+0x507/0x12a0
tty_ioctl+0x355/0xc30
do_vfs_ioctl+0xa3/0x5e0
SyS_ioctl+0x79/0x90
entry_SYSCALL_64_fastpath+0x13/0x94
- i915 unpin_work workqueue:
intel_unpin_work_fn+0x58/0x140 [i915]
process_one_work+0x1f1/0x480
worker_thread+0x48/0x4d0
kthread+0x101/0x140
and this patch purely papers over the issue by adding a NULL pointer
check and a WARN_ON_ONCE() to avoid the oops that would then generally
make the machine unresponsive.
Other callers of i915_gem_object_to_ggtt() seem to also check for the
returned pointer being NULL and warn about it, so this clearly has
happened before in other places.
[ Reported it originally to the i915 developers on Jan 8, applying the
ugly workaround on my own now after triggering the problem for the
second time with no feedback.
This is likely to be the same bug reported as
https://bugs.freedesktop.org/show_bug.cgi?id=98829
https://bugs.freedesktop.org/show_bug.cgi?id=99134
which has a patch for the underlying problem, but it hasn't gotten to
me, so I'm applying the workaround. ]
Cc: Daniel Vetter <daniel.vetter@intel.com>
Cc: Jani Nikula <jani.nikula@linux.intel.com>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Imre Deak <imre.deak@intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>